1 DARWIN Univariate Options Analysis

1.0.1 Loading the libraries

library("FRESA.CAD")
library(readxl)
library("whitening")
library("fpc")
library("robust")
op <- par(no.readonly = TRUE)
pander::panderOptions('digits', 3)
pander::panderOptions('table.split.table', 400)
pander::panderOptions('keep.trailing.zeros',TRUE)
TrainFraction <- 0.50;

dataLoad = FALSE

1.1 The Data

DARWIN <- read.csv("~/GitHub/FCA/Data/DARWIN/DARWIN.csv")
rownames(DARWIN) <- DARWIN$ID
DARWIN$ID <- NULL
DARWIN$class <- 1*(DARWIN$class=="P")
print(table(DARWIN$class))
#> 
#>  0  1 
#> 85 89

DARWIN[,1:ncol(DARWIN)] <- sapply(DARWIN,as.numeric)


set.seed(1)
trainCases <- sample(nrow(DARWIN),nrow(DARWIN)*TrainFraction)

DARWIN_train <- DARWIN[trainCases,]
DARWIN_test <- DARWIN[-trainCases,]

pander::pander(table(DARWIN_train$class))
0 1
48 39
pander::pander(table(DARWIN_test$class))
0 1
37 50

1.1.0.1 Analysis parameters

dataframe <- DARWIN_train
dataframeTest <- DARWIN_test
outcome <- "class"

1.1.1 All the run options

sig_pvalue <- 0.05
thr <- c(0.05,0.2,0.4,0.6,0.8,0.95);
method  <- c("fast","pearson","spearman");
#method  <- c("fast")
type <- c("LM","RLM")
#type <- c("LM")
DeOutcome <- c("T_Blind","T_Driven")
corRank <- c(FALSE,TRUE)

#method  <- c("fast","pearson","spearman");
#type <- c("LM","RLM")

1.2 Decorrelation Analysis


if (dataLoad)
{
  load("~/GitHub/LatentBiomarkers/RMD/DARWIN_ALL_Options.RData")
} else
{
  
  idx = 0;
  thenames <- list();
  totBaM <- NULL
  totDeM <- NULL
  toUnmatM <- NULL
  unalteredM <- NULL
  Decorrleated_FractionM<- NULL
  Base_FractionM<- NULL
  Unaltered_FractionM <- NULL
  sparcityM <- NULL
  Average_Latent_SizeM <- NULL
  SigDeM <- NULL
  La_SignificantM <- NULL
  pbKNNaucM <- NULL
  pbKNNaccM <- NULL
  
  for (DeOut in DeOutcome)
  {
    for (meth in method)
    {
      for (typ in type)
      {
        for (corran in corRank)
        {
          par(op)
          par(mfrow=c(3,2),cex=0.5)
          idx <- idx + 1;
          thenames[[idx]] <- paste(DeOut,meth,typ,corran,sep="_")
          totBa <- numeric()
          totDe <- numeric()
          toUnmat <- numeric()
          unaltered <- numeric()
          Decorrleated_Fraction<- numeric()
          Base_Fraction<- numeric()
          Unaltered_Fraction <- numeric()
          sparcity <- numeric()
          Average_Latent_Size  <- numeric()
          SigDe <- numeric()
          La_Significant <- numeric()
          pbKNNauc <- numeric()
          pbKNNacc <- numeric()
          for (thrs in thr)
          {
            print(thenames[[idx]])
            plotname <- paste(thenames[[idx]],thrs,sep="_")
            
  
            if (DeOut != "T_Driven")
            {
              DEdataframeTrain <- IDeA(dataframe,
                                                     thr=thrs,
                                                     method=meth,
                                                     type=typ,
                                                     corRank=corran,
                                                     relaxed=TRUE,
                                                     verbose = FALSE)
            } else
            {
              DEdataframeTrain <- IDeA(dataframe,
                                                     Outcome=outcome,
                                                     thr=thrs,
                                                     method=meth,
                                                     type=typ,
                                                     corRank=corran,
                                                     relaxed=TRUE,
                                                     verbose = FALSE)
            }
            demat <- attr(DEdataframeTrain,"UPSTM")
            DEdataframe <- predictDecorrelate(DEdataframeTrain,dataframeTest)
            totFe <- ncol(DEdataframe)-1
            totBa <- c(totBa,length(attr(DEdataframeTrain,"unaltered")))
            totDe <- c(totDe, sum(str_detect(colnames(DEdataframe),"La_")))
            toUn <- sum(apply(demat!=0,2,sum)==1)
            toUnmat <- c(toUnmat,toUn )
            una <-  totFe - ncol(demat) + toUn
            unaltered <- c(unaltered,una)
            Decorrleated_Fraction <- c(Decorrleated_Fraction,sum(str_detect(colnames(DEdataframe),"La_"))/totFe)
            Base_Fraction <- c(Base_Fraction,length(attr(DEdataframeTrain,"unaltered"))/totFe)
            
            Unaltered_Fraction <- c(Unaltered_Fraction,una/totFe)
            sparcity <- c(sparcity,(totFe-ncol(demat)+sum(abs(demat)!=0))/totFe/totFe)
            
            varlistDe <-  colnames(demat)[apply(demat!=0,2,sum)>1];
            varlistDe <- as.data.frame(cbind(name=varlistDe,desc=varlistDe))
            
            
            varlist_DeAll <- colnames(DEdataframe)
            varlist_DeAll <- varlist_DeAll[!(varlist_DeAll %in% c(outcome))]
            varlist_DeAll <- as.data.frame(cbind(name=varlist_DeAll,desc=varlist_DeAll))
            
  
            
            pDe <- univariate_Wilcoxon(DEdataframeTrain,outcome,pvalue=sig_pvalue);
            topDecorNames <- names(pDe);
  
            dc <- getLatentCoefficients(DEdataframeTrain)
            deNames_in_dc <- topDecorNames[topDecorNames %in% names(dc)]
            selectedlist <- dc[deNames_in_dc]
            theDeFormulas <- selectedlist
            
            
            Average_Latent_Size <- c(Average_Latent_Size,length(unlist(theDeFormulas))/length(theDeFormulas))
            
            topSigDe <- topDecorNames
            
            pDeDe <- pDe[names(pDe) %in% varlistDe[,1]]

            La_Significant <- c(La_Significant,length(topSigDe))
            SigDe <- c(SigDe,length(pDeDe))
  
            
            mlKNN <- KNN_method(formula(paste(outcome,"~.")),DEdataframeTrain[,c(outcome,topSigDe)])
            psb <- predictionStats_binary(cbind(dataframeTest[,outcome],
                                                  predict(mlKNN,
                                                  DEdataframe[,c(outcome,topSigDe)])),plotname,cex=0.6)
            pbKNNauc <- c(pbKNNauc,psb$aucs[1])
            pbKNNacc <- c(pbKNNacc,psb$accc[1])
          }
          totBaM <- rbind(totBaM,totBa)
          totDeM <- rbind(totDeM,totDe)
          toUnmatM <- rbind(toUnmatM,toUnmat)
          unalteredM <- rbind(unalteredM,unaltered)
          Decorrleated_FractionM <- rbind(Decorrleated_FractionM,Decorrleated_Fraction)
          Base_FractionM <- rbind(Base_FractionM,Base_Fraction)
          Unaltered_FractionM <- rbind(Unaltered_FractionM,Unaltered_Fraction)
          sparcityM <- rbind(sparcityM,sparcity)
          Average_Latent_SizeM <- rbind(Average_Latent_SizeM,Average_Latent_Size)
          SigDeM <- rbind(SigDeM,SigDe)
          La_SignificantM <- rbind(La_SignificantM,La_Significant)
          pbKNNaucM <- rbind(pbKNNaucM,pbKNNauc)
          pbKNNaccM <- rbind(pbKNNaccM,pbKNNacc)
  
        }
      }
    }
  }
}

[1] “T_Blind_fast_LM_FALSE” T_Blind_fast_LM_FALSE_0.05 [1] “T_Blind_fast_LM_FALSE” T_Blind_fast_LM_FALSE_0.2 [1] “T_Blind_fast_LM_FALSE” T_Blind_fast_LM_FALSE_0.4 [1] “T_Blind_fast_LM_FALSE” T_Blind_fast_LM_FALSE_0.6 [1] “T_Blind_fast_LM_FALSE” T_Blind_fast_LM_FALSE_0.8 [1] “T_Blind_fast_LM_FALSE” T_Blind_fast_LM_FALSE_0.95 [1] “T_Blind_fast_LM_TRUE” T_Blind_fast_LM_TRUE_0.05 [1] “T_Blind_fast_LM_TRUE” T_Blind_fast_LM_TRUE_0.2 [1] “T_Blind_fast_LM_TRUE” T_Blind_fast_LM_TRUE_0.4 [1] “T_Blind_fast_LM_TRUE” T_Blind_fast_LM_TRUE_0.6 [1] “T_Blind_fast_LM_TRUE” T_Blind_fast_LM_TRUE_0.8 [1] “T_Blind_fast_LM_TRUE” T_Blind_fast_LM_TRUE_0.95 [1] “T_Blind_fast_RLM_FALSE” T_Blind_fast_RLM_FALSE_0.05 [1] “T_Blind_fast_RLM_FALSE” T_Blind_fast_RLM_FALSE_0.2 [1] “T_Blind_fast_RLM_FALSE” T_Blind_fast_RLM_FALSE_0.4 [1] “T_Blind_fast_RLM_FALSE” T_Blind_fast_RLM_FALSE_0.6 [1] “T_Blind_fast_RLM_FALSE” T_Blind_fast_RLM_FALSE_0.8 [1] “T_Blind_fast_RLM_FALSE” T_Blind_fast_RLM_FALSE_0.95 [1] “T_Blind_fast_RLM_TRUE” T_Blind_fast_RLM_TRUE_0.05 [1] “T_Blind_fast_RLM_TRUE” T_Blind_fast_RLM_TRUE_0.2 [1] “T_Blind_fast_RLM_TRUE” T_Blind_fast_RLM_TRUE_0.4 [1] “T_Blind_fast_RLM_TRUE” T_Blind_fast_RLM_TRUE_0.6 [1] “T_Blind_fast_RLM_TRUE” T_Blind_fast_RLM_TRUE_0.8 [1] “T_Blind_fast_RLM_TRUE” T_Blind_fast_RLM_TRUE_0.95 [1] “T_Blind_pearson_LM_FALSE” T_Blind_pearson_LM_FALSE_0.05 [1] “T_Blind_pearson_LM_FALSE” T_Blind_pearson_LM_FALSE_0.2 [1] “T_Blind_pearson_LM_FALSE” T_Blind_pearson_LM_FALSE_0.4 [1] “T_Blind_pearson_LM_FALSE” T_Blind_pearson_LM_FALSE_0.6 [1] “T_Blind_pearson_LM_FALSE” T_Blind_pearson_LM_FALSE_0.8 [1] “T_Blind_pearson_LM_FALSE” T_Blind_pearson_LM_FALSE_0.95 [1] “T_Blind_pearson_LM_TRUE” T_Blind_pearson_LM_TRUE_0.05 [1] “T_Blind_pearson_LM_TRUE” T_Blind_pearson_LM_TRUE_0.2 [1] “T_Blind_pearson_LM_TRUE” T_Blind_pearson_LM_TRUE_0.4 [1] “T_Blind_pearson_LM_TRUE” T_Blind_pearson_LM_TRUE_0.6 [1] “T_Blind_pearson_LM_TRUE” T_Blind_pearson_LM_TRUE_0.8 [1] “T_Blind_pearson_LM_TRUE” T_Blind_pearson_LM_TRUE_0.95 [1] “T_Blind_pearson_RLM_FALSE” T_Blind_pearson_RLM_FALSE_0.05 [1] “T_Blind_pearson_RLM_FALSE” T_Blind_pearson_RLM_FALSE_0.2 [1] “T_Blind_pearson_RLM_FALSE” T_Blind_pearson_RLM_FALSE_0.4 [1] “T_Blind_pearson_RLM_FALSE” T_Blind_pearson_RLM_FALSE_0.6 [1] “T_Blind_pearson_RLM_FALSE” T_Blind_pearson_RLM_FALSE_0.8 [1] “T_Blind_pearson_RLM_FALSE” T_Blind_pearson_RLM_FALSE_0.95 [1] “T_Blind_pearson_RLM_TRUE” T_Blind_pearson_RLM_TRUE_0.05 [1] “T_Blind_pearson_RLM_TRUE” T_Blind_pearson_RLM_TRUE_0.2 [1] “T_Blind_pearson_RLM_TRUE” T_Blind_pearson_RLM_TRUE_0.4 [1] “T_Blind_pearson_RLM_TRUE” T_Blind_pearson_RLM_TRUE_0.6 [1] “T_Blind_pearson_RLM_TRUE” T_Blind_pearson_RLM_TRUE_0.8 [1] “T_Blind_pearson_RLM_TRUE” T_Blind_pearson_RLM_TRUE_0.95 [1] “T_Blind_spearman_LM_FALSE” T_Blind_spearman_LM_FALSE_0.05 [1] “T_Blind_spearman_LM_FALSE” T_Blind_spearman_LM_FALSE_0.2 [1] “T_Blind_spearman_LM_FALSE” T_Blind_spearman_LM_FALSE_0.4 [1] “T_Blind_spearman_LM_FALSE” T_Blind_spearman_LM_FALSE_0.6 [1] “T_Blind_spearman_LM_FALSE” T_Blind_spearman_LM_FALSE_0.8 [1] “T_Blind_spearman_LM_FALSE” T_Blind_spearman_LM_FALSE_0.95 [1] “T_Blind_spearman_LM_TRUE” T_Blind_spearman_LM_TRUE_0.05 [1] “T_Blind_spearman_LM_TRUE” T_Blind_spearman_LM_TRUE_0.2 [1] “T_Blind_spearman_LM_TRUE” T_Blind_spearman_LM_TRUE_0.4 [1] “T_Blind_spearman_LM_TRUE” T_Blind_spearman_LM_TRUE_0.6 [1] “T_Blind_spearman_LM_TRUE” T_Blind_spearman_LM_TRUE_0.8 [1] “T_Blind_spearman_LM_TRUE” T_Blind_spearman_LM_TRUE_0.95 [1] “T_Blind_spearman_RLM_FALSE” T_Blind_spearman_RLM_FALSE_0.05 [1] “T_Blind_spearman_RLM_FALSE” T_Blind_spearman_RLM_FALSE_0.2 [1] “T_Blind_spearman_RLM_FALSE” T_Blind_spearman_RLM_FALSE_0.4 [1] “T_Blind_spearman_RLM_FALSE” T_Blind_spearman_RLM_FALSE_0.6 [1] “T_Blind_spearman_RLM_FALSE” T_Blind_spearman_RLM_FALSE_0.8 [1] “T_Blind_spearman_RLM_FALSE” T_Blind_spearman_RLM_FALSE_0.95 [1] “T_Blind_spearman_RLM_TRUE” T_Blind_spearman_RLM_TRUE_0.05 [1] “T_Blind_spearman_RLM_TRUE” T_Blind_spearman_RLM_TRUE_0.2 [1] “T_Blind_spearman_RLM_TRUE” T_Blind_spearman_RLM_TRUE_0.4 [1] “T_Blind_spearman_RLM_TRUE” T_Blind_spearman_RLM_TRUE_0.6 [1] “T_Blind_spearman_RLM_TRUE” T_Blind_spearman_RLM_TRUE_0.8 [1] “T_Blind_spearman_RLM_TRUE” T_Blind_spearman_RLM_TRUE_0.95 [1] “T_Driven_fast_LM_FALSE” T_Driven_fast_LM_FALSE_0.05 [1] “T_Driven_fast_LM_FALSE” T_Driven_fast_LM_FALSE_0.2 [1] “T_Driven_fast_LM_FALSE” T_Driven_fast_LM_FALSE_0.4 [1] “T_Driven_fast_LM_FALSE” T_Driven_fast_LM_FALSE_0.6 [1] “T_Driven_fast_LM_FALSE” T_Driven_fast_LM_FALSE_0.8 [1] “T_Driven_fast_LM_FALSE” T_Driven_fast_LM_FALSE_0.95 [1] “T_Driven_fast_LM_TRUE” T_Driven_fast_LM_TRUE_0.05 [1] “T_Driven_fast_LM_TRUE” T_Driven_fast_LM_TRUE_0.2 [1] “T_Driven_fast_LM_TRUE” T_Driven_fast_LM_TRUE_0.4 [1] “T_Driven_fast_LM_TRUE” T_Driven_fast_LM_TRUE_0.6 [1] “T_Driven_fast_LM_TRUE” T_Driven_fast_LM_TRUE_0.8 [1] “T_Driven_fast_LM_TRUE” T_Driven_fast_LM_TRUE_0.95 [1] “T_Driven_fast_RLM_FALSE” T_Driven_fast_RLM_FALSE_0.05 [1] “T_Driven_fast_RLM_FALSE” T_Driven_fast_RLM_FALSE_0.2 [1] “T_Driven_fast_RLM_FALSE” T_Driven_fast_RLM_FALSE_0.4 [1] “T_Driven_fast_RLM_FALSE” T_Driven_fast_RLM_FALSE_0.6 [1] “T_Driven_fast_RLM_FALSE” T_Driven_fast_RLM_FALSE_0.8 [1] “T_Driven_fast_RLM_FALSE” T_Driven_fast_RLM_FALSE_0.95 [1] “T_Driven_fast_RLM_TRUE” T_Driven_fast_RLM_TRUE_0.05 [1] “T_Driven_fast_RLM_TRUE” T_Driven_fast_RLM_TRUE_0.2 [1] “T_Driven_fast_RLM_TRUE” T_Driven_fast_RLM_TRUE_0.4 [1] “T_Driven_fast_RLM_TRUE” T_Driven_fast_RLM_TRUE_0.6 [1] “T_Driven_fast_RLM_TRUE” T_Driven_fast_RLM_TRUE_0.8 [1] “T_Driven_fast_RLM_TRUE” T_Driven_fast_RLM_TRUE_0.95 [1] “T_Driven_pearson_LM_FALSE” T_Driven_pearson_LM_FALSE_0.05 [1] “T_Driven_pearson_LM_FALSE” T_Driven_pearson_LM_FALSE_0.2 [1] “T_Driven_pearson_LM_FALSE” T_Driven_pearson_LM_FALSE_0.4 [1] “T_Driven_pearson_LM_FALSE” T_Driven_pearson_LM_FALSE_0.6 [1] “T_Driven_pearson_LM_FALSE” T_Driven_pearson_LM_FALSE_0.8 [1] “T_Driven_pearson_LM_FALSE” T_Driven_pearson_LM_FALSE_0.95 [1] “T_Driven_pearson_LM_TRUE” T_Driven_pearson_LM_TRUE_0.05 [1] “T_Driven_pearson_LM_TRUE” T_Driven_pearson_LM_TRUE_0.2 [1] “T_Driven_pearson_LM_TRUE” T_Driven_pearson_LM_TRUE_0.4 [1] “T_Driven_pearson_LM_TRUE” T_Driven_pearson_LM_TRUE_0.6 [1] “T_Driven_pearson_LM_TRUE” T_Driven_pearson_LM_TRUE_0.8 [1] “T_Driven_pearson_LM_TRUE” T_Driven_pearson_LM_TRUE_0.95 [1] “T_Driven_pearson_RLM_FALSE” T_Driven_pearson_RLM_FALSE_0.05 [1] “T_Driven_pearson_RLM_FALSE” T_Driven_pearson_RLM_FALSE_0.2 [1] “T_Driven_pearson_RLM_FALSE” T_Driven_pearson_RLM_FALSE_0.4 [1] “T_Driven_pearson_RLM_FALSE” T_Driven_pearson_RLM_FALSE_0.6 [1] “T_Driven_pearson_RLM_FALSE” T_Driven_pearson_RLM_FALSE_0.8 [1] “T_Driven_pearson_RLM_FALSE” T_Driven_pearson_RLM_FALSE_0.95 [1] “T_Driven_pearson_RLM_TRUE” T_Driven_pearson_RLM_TRUE_0.05 [1] “T_Driven_pearson_RLM_TRUE” T_Driven_pearson_RLM_TRUE_0.2 [1] “T_Driven_pearson_RLM_TRUE” T_Driven_pearson_RLM_TRUE_0.4 [1] “T_Driven_pearson_RLM_TRUE” T_Driven_pearson_RLM_TRUE_0.6 [1] “T_Driven_pearson_RLM_TRUE” T_Driven_pearson_RLM_TRUE_0.8 [1] “T_Driven_pearson_RLM_TRUE” T_Driven_pearson_RLM_TRUE_0.95 [1] “T_Driven_spearman_LM_FALSE” T_Driven_spearman_LM_FALSE_0.05 [1] “T_Driven_spearman_LM_FALSE” T_Driven_spearman_LM_FALSE_0.2 [1] “T_Driven_spearman_LM_FALSE” T_Driven_spearman_LM_FALSE_0.4 [1] “T_Driven_spearman_LM_FALSE” T_Driven_spearman_LM_FALSE_0.6 [1] “T_Driven_spearman_LM_FALSE” T_Driven_spearman_LM_FALSE_0.8 [1] “T_Driven_spearman_LM_FALSE” T_Driven_spearman_LM_FALSE_0.95 [1] “T_Driven_spearman_LM_TRUE” T_Driven_spearman_LM_TRUE_0.05 [1] “T_Driven_spearman_LM_TRUE” T_Driven_spearman_LM_TRUE_0.2 [1] “T_Driven_spearman_LM_TRUE” T_Driven_spearman_LM_TRUE_0.4 [1] “T_Driven_spearman_LM_TRUE” T_Driven_spearman_LM_TRUE_0.6 [1] “T_Driven_spearman_LM_TRUE” T_Driven_spearman_LM_TRUE_0.8 [1] “T_Driven_spearman_LM_TRUE” T_Driven_spearman_LM_TRUE_0.95 [1] “T_Driven_spearman_RLM_FALSE” T_Driven_spearman_RLM_FALSE_0.05 [1] “T_Driven_spearman_RLM_FALSE” T_Driven_spearman_RLM_FALSE_0.2 [1] “T_Driven_spearman_RLM_FALSE” T_Driven_spearman_RLM_FALSE_0.4 [1] “T_Driven_spearman_RLM_FALSE” T_Driven_spearman_RLM_FALSE_0.6 [1] “T_Driven_spearman_RLM_FALSE” T_Driven_spearman_RLM_FALSE_0.8 [1] “T_Driven_spearman_RLM_FALSE” T_Driven_spearman_RLM_FALSE_0.95 [1] “T_Driven_spearman_RLM_TRUE” T_Driven_spearman_RLM_TRUE_0.05 [1] “T_Driven_spearman_RLM_TRUE” T_Driven_spearman_RLM_TRUE_0.2 [1] “T_Driven_spearman_RLM_TRUE” T_Driven_spearman_RLM_TRUE_0.4 [1] “T_Driven_spearman_RLM_TRUE” T_Driven_spearman_RLM_TRUE_0.6 [1] “T_Driven_spearman_RLM_TRUE” T_Driven_spearman_RLM_TRUE_0.8 [1] “T_Driven_spearman_RLM_TRUE” T_Driven_spearman_RLM_TRUE_0.95

par(op)

1.2.1 Printing the analysis outputs

par(op)
par(mfrow=c(1,2),cex=0.6)

rownames(totBaM) <- thenames
rownames(totDeM) <- thenames
rownames(toUnmatM) <- thenames
rownames(unalteredM) <- thenames
rownames(Decorrleated_FractionM) <- thenames
rownames(Base_FractionM) <- thenames
rownames(Unaltered_FractionM) <- thenames
rownames(sparcityM) <- thenames
rownames(Average_Latent_SizeM) <- thenames
rownames(SigDeM) <- thenames
rownames(La_SignificantM) <- thenames
rownames(pbKNNaucM) <- thenames
rownames(pbKNNaccM) <- thenames

colnames(totBaM) <- thr
colnames(totDeM) <- thr
colnames(toUnmatM) <- thr
colnames(unalteredM) <- thr
colnames(Decorrleated_FractionM) <- thr
colnames(Base_FractionM) <- thr
colnames(Unaltered_FractionM) <- thr
colnames(sparcityM) <- thr
colnames(Average_Latent_SizeM) <- thr
colnames(SigDeM) <- thr
colnames(La_SignificantM) <- thr
colnames(pbKNNaucM) <- thr
colnames(pbKNNaccM) <- thr

pander::pander(totFe)

450


pander::pander(totBaM)
  0.05 0.2 0.4 0.6 0.8 0.95
T_Blind_fast_LM_FALSE 72 58 68 127 142 83
T_Blind_fast_LM_TRUE 48 20 27 96 127 83
T_Blind_fast_RLM_FALSE 72 58 68 127 142 83
T_Blind_fast_RLM_TRUE 48 20 27 96 127 83
T_Blind_pearson_LM_FALSE 55 35 71 128 142 83
T_Blind_pearson_LM_TRUE 37 23 23 98 127 83
T_Blind_pearson_RLM_FALSE 89 81 90 129 138 82
T_Blind_pearson_RLM_TRUE 75 72 65 88 124 82
T_Blind_spearman_LM_FALSE 68 69 60 121 137 52
T_Blind_spearman_LM_TRUE 54 55 19 94 119 52
T_Blind_spearman_RLM_FALSE 83 78 72 118 140 50
T_Blind_spearman_RLM_TRUE 73 75 43 94 122 50
T_Driven_fast_LM_FALSE 95 44 61 126 138 83
T_Driven_fast_LM_TRUE 66 60 38 100 127 83
T_Driven_fast_RLM_FALSE 95 44 61 126 138 83
T_Driven_fast_RLM_TRUE 66 60 38 100 127 83
T_Driven_pearson_LM_FALSE 52 61 61 125 138 83
T_Driven_pearson_LM_TRUE 53 48 40 99 127 83
T_Driven_pearson_RLM_FALSE 92 93 95 125 134 80
T_Driven_pearson_RLM_TRUE 83 73 73 103 123 80
T_Driven_spearman_LM_FALSE 67 60 67 119 134 51
T_Driven_spearman_LM_TRUE 66 54 30 96 123 51
T_Driven_spearman_RLM_FALSE 84 68 76 115 134 50
T_Driven_spearman_RLM_TRUE 76 76 65 98 125 50
pander::pander(totDeM)
  0.05 0.2 0.4 0.6 0.8 0.95
T_Blind_fast_LM_FALSE 357 377 377 276 178 78
T_Blind_fast_LM_TRUE 380 422 421 306 194 79
T_Blind_fast_RLM_FALSE 357 377 377 276 178 78
T_Blind_fast_RLM_TRUE 380 422 421 306 194 79
T_Blind_pearson_LM_FALSE 381 407 367 276 178 78
T_Blind_pearson_LM_TRUE 390 414 424 305 194 79
T_Blind_pearson_RLM_FALSE 314 324 315 264 175 78
T_Blind_pearson_RLM_TRUE 334 335 339 290 189 79
T_Blind_spearman_LM_FALSE 362 361 383 280 175 52
T_Blind_spearman_LM_TRUE 370 368 428 316 192 52
T_Blind_spearman_RLM_FALSE 331 338 357 273 172 52
T_Blind_spearman_RLM_TRUE 342 341 394 315 190 52
T_Driven_fast_LM_FALSE 327 391 383 282 178 78
T_Driven_fast_LM_TRUE 357 363 410 306 190 79
T_Driven_fast_RLM_FALSE 327 391 383 282 178 78
T_Driven_fast_RLM_TRUE 357 363 410 306 190 79
T_Driven_pearson_LM_FALSE 375 370 384 282 178 78
T_Driven_pearson_LM_TRUE 371 379 408 306 190 79
T_Driven_pearson_RLM_FALSE 306 309 313 266 174 78
T_Driven_pearson_RLM_TRUE 323 329 327 284 186 79
T_Driven_spearman_LM_FALSE 357 365 373 288 174 52
T_Driven_spearman_LM_TRUE 357 369 412 314 184 52
T_Driven_spearman_RLM_FALSE 329 349 348 284 171 52
T_Driven_spearman_RLM_TRUE 342 338 369 308 182 52
pander::pander(toUnmatM)
  0.05 0.2 0.4 0.6 0.8 0.95
T_Blind_fast_LM_FALSE 72 58 68 127 142 83
T_Blind_fast_LM_TRUE 48 20 27 96 127 83
T_Blind_fast_RLM_FALSE 72 58 68 127 142 83
T_Blind_fast_RLM_TRUE 48 20 27 96 127 83
T_Blind_pearson_LM_FALSE 55 35 71 128 142 83
T_Blind_pearson_LM_TRUE 37 23 23 98 127 83
T_Blind_pearson_RLM_FALSE 89 81 90 129 138 82
T_Blind_pearson_RLM_TRUE 75 72 65 88 124 82
T_Blind_spearman_LM_FALSE 68 69 60 121 137 52
T_Blind_spearman_LM_TRUE 54 55 19 94 119 52
T_Blind_spearman_RLM_FALSE 83 78 72 118 140 50
T_Blind_spearman_RLM_TRUE 73 75 43 94 122 50
T_Driven_fast_LM_FALSE 95 44 61 126 138 83
T_Driven_fast_LM_TRUE 66 60 38 100 127 83
T_Driven_fast_RLM_FALSE 95 44 61 126 138 83
T_Driven_fast_RLM_TRUE 66 60 38 100 127 83
T_Driven_pearson_LM_FALSE 52 61 61 125 138 83
T_Driven_pearson_LM_TRUE 53 48 40 99 127 83
T_Driven_pearson_RLM_FALSE 92 93 95 125 134 80
T_Driven_pearson_RLM_TRUE 83 73 73 103 123 80
T_Driven_spearman_LM_FALSE 67 60 67 119 134 51
T_Driven_spearman_LM_TRUE 66 54 30 96 123 51
T_Driven_spearman_RLM_FALSE 84 68 76 115 134 50
T_Driven_spearman_RLM_TRUE 76 76 65 98 125 50
pander::pander(unalteredM)
  0.05 0.2 0.4 0.6 0.8 0.95
T_Blind_fast_LM_FALSE 93 73 73 174 272 372
T_Blind_fast_LM_TRUE 70 28 29 144 256 371
T_Blind_fast_RLM_FALSE 93 73 73 174 272 372
T_Blind_fast_RLM_TRUE 70 28 29 144 256 371
T_Blind_pearson_LM_FALSE 69 43 83 174 272 372
T_Blind_pearson_LM_TRUE 60 36 26 145 256 371
T_Blind_pearson_RLM_FALSE 136 126 135 186 275 372
T_Blind_pearson_RLM_TRUE 116 115 111 160 261 371
T_Blind_spearman_LM_FALSE 88 89 67 170 275 398
T_Blind_spearman_LM_TRUE 80 82 22 134 258 398
T_Blind_spearman_RLM_FALSE 119 112 93 177 278 398
T_Blind_spearman_RLM_TRUE 108 109 56 135 260 398
T_Driven_fast_LM_FALSE 123 59 67 168 272 372
T_Driven_fast_LM_TRUE 93 87 40 144 260 371
T_Driven_fast_RLM_FALSE 123 59 67 168 272 372
T_Driven_fast_RLM_TRUE 93 87 40 144 260 371
T_Driven_pearson_LM_FALSE 75 80 66 168 272 372
T_Driven_pearson_LM_TRUE 79 71 42 144 260 371
T_Driven_pearson_RLM_FALSE 144 141 137 184 276 372
T_Driven_pearson_RLM_TRUE 127 121 123 166 264 371
T_Driven_spearman_LM_FALSE 93 85 77 162 276 398
T_Driven_spearman_LM_TRUE 93 81 38 136 266 398
T_Driven_spearman_RLM_FALSE 121 101 102 166 279 398
T_Driven_spearman_RLM_TRUE 108 112 81 142 268 398
pander::pander(Decorrleated_FractionM)
  0.05 0.2 0.4 0.6 0.8 0.95
T_Blind_fast_LM_FALSE 0.793 0.838 0.838 0.613 0.396 0.173
T_Blind_fast_LM_TRUE 0.844 0.938 0.936 0.680 0.431 0.176
T_Blind_fast_RLM_FALSE 0.793 0.838 0.838 0.613 0.396 0.173
T_Blind_fast_RLM_TRUE 0.844 0.938 0.936 0.680 0.431 0.176
T_Blind_pearson_LM_FALSE 0.847 0.904 0.816 0.613 0.396 0.173
T_Blind_pearson_LM_TRUE 0.867 0.920 0.942 0.678 0.431 0.176
T_Blind_pearson_RLM_FALSE 0.698 0.720 0.700 0.587 0.389 0.173
T_Blind_pearson_RLM_TRUE 0.742 0.744 0.753 0.644 0.420 0.176
T_Blind_spearman_LM_FALSE 0.804 0.802 0.851 0.622 0.389 0.116
T_Blind_spearman_LM_TRUE 0.822 0.818 0.951 0.702 0.427 0.116
T_Blind_spearman_RLM_FALSE 0.736 0.751 0.793 0.607 0.382 0.116
T_Blind_spearman_RLM_TRUE 0.760 0.758 0.876 0.700 0.422 0.116
T_Driven_fast_LM_FALSE 0.727 0.869 0.851 0.627 0.396 0.173
T_Driven_fast_LM_TRUE 0.793 0.807 0.911 0.680 0.422 0.176
T_Driven_fast_RLM_FALSE 0.727 0.869 0.851 0.627 0.396 0.173
T_Driven_fast_RLM_TRUE 0.793 0.807 0.911 0.680 0.422 0.176
T_Driven_pearson_LM_FALSE 0.833 0.822 0.853 0.627 0.396 0.173
T_Driven_pearson_LM_TRUE 0.824 0.842 0.907 0.680 0.422 0.176
T_Driven_pearson_RLM_FALSE 0.680 0.687 0.696 0.591 0.387 0.173
T_Driven_pearson_RLM_TRUE 0.718 0.731 0.727 0.631 0.413 0.176
T_Driven_spearman_LM_FALSE 0.793 0.811 0.829 0.640 0.387 0.116
T_Driven_spearman_LM_TRUE 0.793 0.820 0.916 0.698 0.409 0.116
T_Driven_spearman_RLM_FALSE 0.731 0.776 0.773 0.631 0.380 0.116
T_Driven_spearman_RLM_TRUE 0.760 0.751 0.820 0.684 0.404 0.116
pander::pander(Base_FractionM)
  0.05 0.2 0.4 0.6 0.8 0.95
T_Blind_fast_LM_FALSE 0.1600 0.1289 0.1511 0.282 0.316 0.184
T_Blind_fast_LM_TRUE 0.1067 0.0444 0.0600 0.213 0.282 0.184
T_Blind_fast_RLM_FALSE 0.1600 0.1289 0.1511 0.282 0.316 0.184
T_Blind_fast_RLM_TRUE 0.1067 0.0444 0.0600 0.213 0.282 0.184
T_Blind_pearson_LM_FALSE 0.1222 0.0778 0.1578 0.284 0.316 0.184
T_Blind_pearson_LM_TRUE 0.0822 0.0511 0.0511 0.218 0.282 0.184
T_Blind_pearson_RLM_FALSE 0.1978 0.1800 0.2000 0.287 0.307 0.182
T_Blind_pearson_RLM_TRUE 0.1667 0.1600 0.1444 0.196 0.276 0.182
T_Blind_spearman_LM_FALSE 0.1511 0.1533 0.1333 0.269 0.304 0.116
T_Blind_spearman_LM_TRUE 0.1200 0.1222 0.0422 0.209 0.264 0.116
T_Blind_spearman_RLM_FALSE 0.1844 0.1733 0.1600 0.262 0.311 0.111
T_Blind_spearman_RLM_TRUE 0.1622 0.1667 0.0956 0.209 0.271 0.111
T_Driven_fast_LM_FALSE 0.2111 0.0978 0.1356 0.280 0.307 0.184
T_Driven_fast_LM_TRUE 0.1467 0.1333 0.0844 0.222 0.282 0.184
T_Driven_fast_RLM_FALSE 0.2111 0.0978 0.1356 0.280 0.307 0.184
T_Driven_fast_RLM_TRUE 0.1467 0.1333 0.0844 0.222 0.282 0.184
T_Driven_pearson_LM_FALSE 0.1156 0.1356 0.1356 0.278 0.307 0.184
T_Driven_pearson_LM_TRUE 0.1178 0.1067 0.0889 0.220 0.282 0.184
T_Driven_pearson_RLM_FALSE 0.2044 0.2067 0.2111 0.278 0.298 0.178
T_Driven_pearson_RLM_TRUE 0.1844 0.1622 0.1622 0.229 0.273 0.178
T_Driven_spearman_LM_FALSE 0.1489 0.1333 0.1489 0.264 0.298 0.113
T_Driven_spearman_LM_TRUE 0.1467 0.1200 0.0667 0.213 0.273 0.113
T_Driven_spearman_RLM_FALSE 0.1867 0.1511 0.1689 0.256 0.298 0.111
T_Driven_spearman_RLM_TRUE 0.1689 0.1689 0.1444 0.218 0.278 0.111
pander::pander(Unaltered_FractionM)
  0.05 0.2 0.4 0.6 0.8 0.95
T_Blind_fast_LM_FALSE 0.207 0.1622 0.1622 0.387 0.604 0.827
T_Blind_fast_LM_TRUE 0.156 0.0622 0.0644 0.320 0.569 0.824
T_Blind_fast_RLM_FALSE 0.207 0.1622 0.1622 0.387 0.604 0.827
T_Blind_fast_RLM_TRUE 0.156 0.0622 0.0644 0.320 0.569 0.824
T_Blind_pearson_LM_FALSE 0.153 0.0956 0.1844 0.387 0.604 0.827
T_Blind_pearson_LM_TRUE 0.133 0.0800 0.0578 0.322 0.569 0.824
T_Blind_pearson_RLM_FALSE 0.302 0.2800 0.3000 0.413 0.611 0.827
T_Blind_pearson_RLM_TRUE 0.258 0.2556 0.2467 0.356 0.580 0.824
T_Blind_spearman_LM_FALSE 0.196 0.1978 0.1489 0.378 0.611 0.884
T_Blind_spearman_LM_TRUE 0.178 0.1822 0.0489 0.298 0.573 0.884
T_Blind_spearman_RLM_FALSE 0.264 0.2489 0.2067 0.393 0.618 0.884
T_Blind_spearman_RLM_TRUE 0.240 0.2422 0.1244 0.300 0.578 0.884
T_Driven_fast_LM_FALSE 0.273 0.1311 0.1489 0.373 0.604 0.827
T_Driven_fast_LM_TRUE 0.207 0.1933 0.0889 0.320 0.578 0.824
T_Driven_fast_RLM_FALSE 0.273 0.1311 0.1489 0.373 0.604 0.827
T_Driven_fast_RLM_TRUE 0.207 0.1933 0.0889 0.320 0.578 0.824
T_Driven_pearson_LM_FALSE 0.167 0.1778 0.1467 0.373 0.604 0.827
T_Driven_pearson_LM_TRUE 0.176 0.1578 0.0933 0.320 0.578 0.824
T_Driven_pearson_RLM_FALSE 0.320 0.3133 0.3044 0.409 0.613 0.827
T_Driven_pearson_RLM_TRUE 0.282 0.2689 0.2733 0.369 0.587 0.824
T_Driven_spearman_LM_FALSE 0.207 0.1889 0.1711 0.360 0.613 0.884
T_Driven_spearman_LM_TRUE 0.207 0.1800 0.0844 0.302 0.591 0.884
T_Driven_spearman_RLM_FALSE 0.269 0.2244 0.2267 0.369 0.620 0.884
T_Driven_spearman_RLM_TRUE 0.240 0.2489 0.1800 0.316 0.596 0.884
pander::pander(sparcityM)
  0.05 0.2 0.4 0.6 0.8 0.95
T_Blind_fast_LM_FALSE 0.00526 0.00586 0.00790 0.00416 0.00330 0.00265
T_Blind_fast_LM_TRUE 0.00595 0.00987 0.01249 0.00457 0.00341 0.00266
T_Blind_fast_RLM_FALSE 0.00526 0.00586 0.00790 0.00416 0.00330 0.00265
T_Blind_fast_RLM_TRUE 0.00595 0.00987 0.01249 0.00457 0.00341 0.00266
T_Blind_pearson_LM_FALSE 0.00575 0.00769 0.00621 0.00413 0.00330 0.00265
T_Blind_pearson_LM_TRUE 0.00577 0.00914 0.01491 0.00462 0.00341 0.00266
T_Blind_pearson_RLM_FALSE 0.00448 0.00475 0.00481 0.00403 0.00331 0.00265
T_Blind_pearson_RLM_TRUE 0.00492 0.00487 0.00506 0.00454 0.00341 0.00266
T_Blind_spearman_LM_FALSE 0.00577 0.00577 0.00765 0.00416 0.00324 0.00249
T_Blind_spearman_LM_TRUE 0.00530 0.00531 0.01131 0.00438 0.00333 0.00249
T_Blind_spearman_RLM_FALSE 0.00492 0.00505 0.00574 0.00420 0.00329 0.00248
T_Blind_spearman_RLM_TRUE 0.00493 0.00495 0.00807 0.00449 0.00337 0.00248
T_Driven_fast_LM_FALSE 0.00472 0.00739 0.00828 0.00426 0.00327 0.00265
T_Driven_fast_LM_TRUE 0.00547 0.00575 0.01221 0.00462 0.00336 0.00266
T_Driven_fast_RLM_FALSE 0.00472 0.00739 0.00828 0.00426 0.00327 0.00265
T_Driven_fast_RLM_TRUE 0.00547 0.00575 0.01221 0.00462 0.00336 0.00266
T_Driven_pearson_LM_FALSE 0.00556 0.00561 0.00826 0.00425 0.00327 0.00265
T_Driven_pearson_LM_TRUE 0.00551 0.00587 0.01137 0.00467 0.00335 0.00266
T_Driven_pearson_RLM_FALSE 0.00444 0.00451 0.00460 0.00407 0.00327 0.00264
T_Driven_pearson_RLM_TRUE 0.00480 0.00481 0.00480 0.00442 0.00336 0.00265
T_Driven_spearman_LM_FALSE 0.00538 0.00539 0.00683 0.00417 0.00321 0.00248
T_Driven_spearman_LM_TRUE 0.00514 0.00539 0.01046 0.00446 0.00328 0.00248
T_Driven_spearman_RLM_FALSE 0.00490 0.00518 0.00550 0.00434 0.00322 0.00248
T_Driven_spearman_RLM_TRUE 0.00509 0.00509 0.00650 0.00438 0.00329 0.00248
pander::pander(Average_Latent_SizeM)
  0.05 0.2 0.4 0.6 0.8 0.95
T_Blind_fast_LM_FALSE 3.12 2.63 5.57 2.83 2.25 2.33
T_Blind_fast_LM_TRUE 3.15 5.00 5.20 2.89 2.07 2.33
T_Blind_fast_RLM_FALSE 3.12 2.63 5.57 2.83 2.25 2.33
T_Blind_fast_RLM_TRUE 3.15 5.00 5.20 2.89 2.07 2.33
T_Blind_pearson_LM_FALSE 2.47 3.09 3.00 2.78 2.29 2.40
T_Blind_pearson_LM_TRUE 2.60 5.83 6.44 3.50 2.26 2.40
T_Blind_pearson_RLM_FALSE 2.35 2.52 2.91 2.32 2.35 2.20
T_Blind_pearson_RLM_TRUE 2.71 2.66 2.95 2.35 2.28 2.20
T_Blind_spearman_LM_FALSE 3.40 3.40 4.00 2.63 2.31 2.12
T_Blind_spearman_LM_TRUE 3.00 3.00 5.87 2.57 2.26 2.10
T_Blind_spearman_RLM_FALSE 2.71 2.90 3.12 2.35 2.44 2.00
T_Blind_spearman_RLM_TRUE 2.79 2.58 5.19 2.42 2.33 2.00
T_Driven_fast_LM_FALSE 3.00 2.67 4.00 3.38 2.41 2.38
T_Driven_fast_LM_TRUE 3.08 2.78 8.29 2.80 2.33 2.38
T_Driven_fast_RLM_FALSE 3.00 2.67 4.00 3.38 2.41 2.38
T_Driven_fast_RLM_TRUE 3.08 2.78 8.29 2.80 2.33 2.38
T_Driven_pearson_LM_FALSE 2.65 2.88 4.00 3.33 2.38 2.44
T_Driven_pearson_LM_TRUE 3.00 2.67 8.11 3.00 2.39 2.44
T_Driven_pearson_RLM_FALSE 2.40 2.77 2.92 2.47 2.38 2.33
T_Driven_pearson_RLM_TRUE 2.62 2.52 3.00 2.60 2.21 2.33
T_Driven_spearman_LM_FALSE 3.05 3.00 5.53 2.83 2.28 2.00
T_Driven_spearman_LM_TRUE 2.92 2.89 5.29 2.44 2.26 2.00
T_Driven_spearman_RLM_FALSE 2.73 2.85 2.57 2.76 2.33 2.00
T_Driven_spearman_RLM_TRUE 2.87 3.04 3.29 2.65 2.38 2.00
pander::pander(SigDeM)
  0.05 0.2 0.4 0.6 0.8 0.95
T_Blind_fast_LM_FALSE 17 19 7 6 16 9
T_Blind_fast_LM_TRUE 13 7 5 9 14 9
T_Blind_fast_RLM_FALSE 17 19 7 6 16 9
T_Blind_fast_RLM_TRUE 13 7 5 9 14 9
T_Blind_pearson_LM_FALSE 15 11 12 9 17 10
T_Blind_pearson_LM_TRUE 10 6 9 12 19 10
T_Blind_pearson_RLM_FALSE 31 27 22 19 17 5
T_Blind_pearson_RLM_TRUE 24 29 22 20 18 5
T_Blind_spearman_LM_FALSE 20 20 13 19 16 8
T_Blind_spearman_LM_TRUE 13 10 15 14 19 10
T_Blind_spearman_RLM_FALSE 21 20 16 26 18 7
T_Blind_spearman_RLM_TRUE 19 19 21 26 18 8
T_Driven_fast_LM_FALSE 11 6 11 8 17 8
T_Driven_fast_LM_TRUE 12 9 7 10 15 8
T_Driven_fast_RLM_FALSE 11 6 11 8 17 8
T_Driven_fast_RLM_TRUE 12 9 7 10 15 8
T_Driven_pearson_LM_FALSE 26 8 8 9 16 9
T_Driven_pearson_LM_TRUE 9 6 9 11 18 9
T_Driven_pearson_RLM_FALSE 25 22 12 19 13 6
T_Driven_pearson_RLM_TRUE 24 21 15 15 14 6
T_Driven_spearman_LM_FALSE 21 10 15 12 18 9
T_Driven_spearman_LM_TRUE 13 9 21 9 19 10
T_Driven_spearman_RLM_FALSE 22 20 23 25 12 6
T_Driven_spearman_RLM_TRUE 23 23 24 20 16 7
pander::pander(La_SignificantM)
  0.05 0.2 0.4 0.6 0.8 0.95
T_Blind_fast_LM_FALSE 32 31 16 34 81 123
T_Blind_fast_LM_TRUE 26 7 8 33 71 119
T_Blind_fast_RLM_FALSE 32 31 16 34 81 123
T_Blind_fast_RLM_TRUE 26 7 8 33 71 119
T_Blind_pearson_LM_FALSE 27 17 22 38 83 124
T_Blind_pearson_LM_TRUE 19 7 13 41 77 123
T_Blind_pearson_RLM_FALSE 59 45 44 56 85 113
T_Blind_pearson_RLM_TRUE 42 49 42 50 81 113
T_Blind_spearman_LM_FALSE 29 29 24 47 85 133
T_Blind_spearman_LM_TRUE 26 21 18 32 87 137
T_Blind_spearman_RLM_FALSE 33 32 28 64 89 132
T_Blind_spearman_RLM_TRUE 39 39 27 48 87 133
T_Driven_fast_LM_FALSE 31 15 31 53 98 126
T_Driven_fast_LM_TRUE 30 25 24 51 89 125
T_Driven_fast_RLM_FALSE 31 15 31 53 98 126
T_Driven_fast_RLM_TRUE 30 25 24 51 89 125
T_Driven_pearson_LM_FALSE 33 15 27 54 96 127
T_Driven_pearson_LM_TRUE 18 12 26 54 92 126
T_Driven_pearson_RLM_FALSE 53 38 42 70 94 124
T_Driven_pearson_RLM_TRUE 44 36 42 63 91 123
T_Driven_spearman_LM_FALSE 28 12 30 45 99 138
T_Driven_spearman_LM_TRUE 26 12 28 36 98 140
T_Driven_spearman_RLM_FALSE 35 24 41 66 93 133
T_Driven_spearman_RLM_TRUE 42 36 39 53 96 134
pander::pander(pbKNNaucM)
  0.05 0.2 0.4 0.6 0.8 0.95
T_Blind_fast_LM_FALSE 0.947 0.951 0.928 0.910 0.928 0.936
T_Blind_fast_LM_TRUE 0.921 0.884 0.860 0.885 0.926 0.934
T_Blind_fast_RLM_FALSE 0.947 0.951 0.928 0.910 0.928 0.936
T_Blind_fast_RLM_TRUE 0.921 0.884 0.860 0.885 0.926 0.934
T_Blind_pearson_LM_FALSE 0.876 0.872 0.906 0.910 0.943 0.932
T_Blind_pearson_LM_TRUE 0.840 0.907 0.828 0.905 0.935 0.934
T_Blind_pearson_RLM_FALSE 0.911 0.919 0.925 0.942 0.919 0.954
T_Blind_pearson_RLM_TRUE 0.947 0.954 0.917 0.906 0.884 0.949
T_Blind_spearman_LM_FALSE 0.960 0.960 0.899 0.928 0.946 0.934
T_Blind_spearman_LM_TRUE 0.960 0.939 0.930 0.931 0.943 0.931
T_Blind_spearman_RLM_FALSE 0.913 0.919 0.926 0.921 0.934 0.929
T_Blind_spearman_RLM_TRUE 0.937 0.951 0.837 0.912 0.933 0.921
T_Driven_fast_LM_FALSE 0.945 0.918 0.969 0.931 0.939 0.943
T_Driven_fast_LM_TRUE 0.931 0.909 0.914 0.931 0.945 0.944
T_Driven_fast_RLM_FALSE 0.945 0.918 0.969 0.931 0.939 0.943
T_Driven_fast_RLM_TRUE 0.931 0.909 0.914 0.931 0.945 0.944
T_Driven_pearson_LM_FALSE 0.934 0.900 0.949 0.939 0.939 0.942
T_Driven_pearson_LM_TRUE 0.840 0.889 0.933 0.934 0.932 0.944
T_Driven_pearson_RLM_FALSE 0.876 0.888 0.926 0.936 0.926 0.947
T_Driven_pearson_RLM_TRUE 0.950 0.924 0.947 0.930 0.932 0.947
T_Driven_spearman_LM_FALSE 0.934 0.932 0.910 0.957 0.941 0.939
T_Driven_spearman_LM_TRUE 0.967 0.910 0.938 0.929 0.941 0.939
T_Driven_spearman_RLM_FALSE 0.925 0.920 0.943 0.956 0.916 0.927
T_Driven_spearman_RLM_TRUE 0.933 0.948 0.919 0.955 0.931 0.941
pander::pander(pbKNNaccM)
  0.05 0.2 0.4 0.6 0.8 0.95
T_Blind_fast_LM_FALSE 0.77 0.736 0.805 0.701 0.724 0.724
T_Blind_fast_LM_TRUE 0.759 0.736 0.77 0.678 0.701 0.736
T_Blind_fast_RLM_FALSE 0.77 0.736 0.805 0.701 0.724 0.724
T_Blind_fast_RLM_TRUE 0.759 0.736 0.77 0.678 0.701 0.736
T_Blind_pearson_LM_FALSE 0.759 0.713 0.701 0.736 0.747 0.724
T_Blind_pearson_LM_TRUE 0.713 0.839 0.724 0.747 0.713 0.724
T_Blind_pearson_RLM_FALSE 0.759 0.747 0.77 0.736 0.713 0.724
T_Blind_pearson_RLM_TRUE 0.759 0.793 0.736 0.736 0.678 0.736
T_Blind_spearman_LM_FALSE 0.77 0.77 0.724 0.793 0.759 0.713
T_Blind_spearman_LM_TRUE 0.782 0.747 0.77 0.759 0.736 0.713
T_Blind_spearman_RLM_FALSE 0.747 0.759 0.77 0.793 0.724 0.724
T_Blind_spearman_RLM_TRUE 0.701 0.678 0.724 0.69 0.713 0.724
T_Driven_fast_LM_FALSE 0.759 0.816 0.793 0.77 0.747 0.724
T_Driven_fast_LM_TRUE 0.793 0.782 0.77 0.713 0.747 0.724
T_Driven_fast_RLM_FALSE 0.759 0.816 0.793 0.77 0.747 0.724
T_Driven_fast_RLM_TRUE 0.793 0.782 0.77 0.713 0.747 0.724
T_Driven_pearson_LM_FALSE 0.793 0.759 0.793 0.759 0.724 0.713
T_Driven_pearson_LM_TRUE 0.69 0.759 0.805 0.759 0.747 0.736
T_Driven_pearson_RLM_FALSE 0.713 0.701 0.77 0.724 0.77 0.713
T_Driven_pearson_RLM_TRUE 0.747 0.77 0.747 0.77 0.759 0.713
T_Driven_spearman_LM_FALSE 0.713 0.828 0.759 0.793 0.747 0.724
T_Driven_spearman_LM_TRUE 0.747 0.793 0.713 0.759 0.747 0.69
T_Driven_spearman_RLM_FALSE 0.736 0.759 0.724 0.77 0.736 0.713
T_Driven_spearman_RLM_TRUE 0.77 0.736 0.69 0.782 0.736 0.713

miny = 0.15
maxy = max(pbKNNaucM)

plot(thr,pbKNNaucM[1,],ylim=c(miny,maxy),
     main="KNN's ROCAUC",
     xlab="Correlation-Matrix's Maximum Goal",
     ylab="ROC AUC",
     type="l",
     col=1,
     lwd=2)
for (ind in 2:nrow(pbKNNaucM))
{
  lines(thr,pbKNNaucM[ind,],col=ind,lwd=2,lty=ind)
}

legend("bottomright", rownames(pbKNNaucM),lty=1:length(thenames), col = 1:length(thenames),cex=0.55,ncol=2)

fastRows <- str_detect(rownames(pbKNNaucM),"fast")
pearsonRows <- str_detect(rownames(pbKNNaucM),"pearson")
spearmanRows <- str_detect(rownames(pbKNNaucM),"spearman")
T_BlindRows <- str_detect(rownames(pbKNNaucM),"T_Blind")
corRankRows <- str_detect(rownames(pbKNNaucM),"TRUE")
maxCorRankRows <- str_detect(rownames(pbKNNaucM),"FALSE")
RLMfitMethod <- str_detect(rownames(pbKNNaucM),"RLM")


meanAuc <-  colMeans(pbKNNaucM[fastRows,])
meanAuc <-  rbind(meanAuc,colMeans(pbKNNaucM[pearsonRows,]))
meanAuc <-  rbind(meanAuc,colMeans(pbKNNaucM[spearmanRows,]))
meanAuc <-  rbind(meanAuc,colMeans(pbKNNaucM[!T_BlindRows,]))
meanAuc <-  rbind(meanAuc,colMeans(pbKNNaucM[T_BlindRows,]))
meanAuc <-  rbind(meanAuc,colMeans(pbKNNaucM[corRankRows,]))
meanAuc <-  rbind(meanAuc,colMeans(pbKNNaucM[maxCorRankRows,]))
meanAuc <-  rbind(meanAuc,colMeans(pbKNNaucM[RLMfitMethod,]))
meanAuc <-  rbind(meanAuc,colMeans(pbKNNaucM[!RLMfitMethod,]))
legnames <- c("fast","Pearson","Spearman","T_Driven","T_Blind","SumCor","MaxCor","RLM","LM")


pbKNNaccM <- as.data.frame(pbKNNaccM)
pbKNNaccM[,1:ncol(pbKNNaccM)] <- sapply(pbKNNaccM,as.numeric)

Average_Latent_SizeM <- as.data.frame(Average_Latent_SizeM)
Average_Latent_SizeM[,1:ncol(Average_Latent_SizeM)] <- sapply(Average_Latent_SizeM,as.numeric)
Average_Latent_SizeM[is.na(Average_Latent_SizeM)] <- 0

SigDeM <- as.data.frame(SigDeM)
SigDeM[,1:ncol(SigDeM)] <- sapply(SigDeM,as.numeric)

sparcityM <- as.data.frame(sparcityM)
sparcityM[,1:ncol(sparcityM)] <- sapply(sparcityM,as.numeric)

miny = 0.65
maxy = max(meanAuc)+0.025

plot(thr,meanAuc[1,],ylim=c(miny,maxy),
     main="Mean KNN's ROCAUC",
     xlab="Correlation-Matrix's Maximum Goal",
     ylab="ROC AUC",
     type="l",
     col=1,
     lwd=2,
     lty=1)
for (ind in 2:nrow(meanAuc))
{
  lines(thr,meanAuc[ind,],col=ind,lwd=2,lty=ind)
}
legend("bottomright", legnames,lty=1:length(legnames), col = 1:length(legnames),cex=0.75)



miny = 0.35
maxy = max(pbKNNaccM) + 0.1

plot(thr,pbKNNaccM[1,],ylim=c(miny,maxy),
     main="KNN's Accuracy",
     xlab="Correlation-Matrix's Maximum Goal",
     ylab="Accuracy",
     type="l",
     col=1,
     lwd=2)
for (ind in 2:nrow(pbKNNaucM))
{
  lines(thr,pbKNNaccM[ind,],col=ind,lwd=2,lty=ind)
}
legend("bottomright", rownames(pbKNNaucM),lty=1:length(thenames), col = 1:length(thenames),cex=0.55,ncol=2)



meanAcc <-  colMeans(pbKNNaccM[fastRows,])
meanAcc <-  rbind(meanAcc,colMeans(pbKNNaccM[pearsonRows,]))
meanAcc <-  rbind(meanAcc,colMeans(pbKNNaccM[spearmanRows,]))
meanAcc <-  rbind(meanAcc,colMeans(pbKNNaccM[!T_BlindRows,]))
meanAcc <-  rbind(meanAcc,colMeans(pbKNNaccM[T_BlindRows,]))
meanAcc <-  rbind(meanAcc,colMeans(pbKNNaccM[corRankRows,]))
meanAcc <-  rbind(meanAcc,colMeans(pbKNNaccM[maxCorRankRows,]))
meanAcc <-  rbind(meanAcc,colMeans(pbKNNaccM[RLMfitMethod,]))
meanAcc <-  rbind(meanAcc,colMeans(pbKNNaccM[!RLMfitMethod,]))

miny = min(meanAcc)-0.01
maxy = max(meanAcc)+0.025

plot(thr,meanAcc[1,],ylim=c(miny,maxy),
     main="Mean KNN's Accuracy",
     xlab="Correlation-Matrix's Maximum Goal",
     ylab="Accuracy",
     type="l",
     col=1,
     lwd=2)
for (ind in 2:nrow(meanAcc))
{
  lines(thr,meanAcc[ind,],col=ind,lwd=2,lty=ind)
}
legend("topleft", legnames,lty=1:length(legnames), col = 1:length(legnames),cex=0.75)





miny = 1
maxy = 20*max(Average_Latent_SizeM)

plot(thr,Average_Latent_SizeM[1,],ylim=c(miny,maxy),
     main="Average Size of Latent-Variable",
     xlab="Correlation-Matrix's Maximum Goal",
     ylab="Size",
     type="l",
     col=1,
     lwd=2,
     log = "y")

for (ind in 2:nrow(Average_Latent_SizeM))
{
  lines(thr,Average_Latent_SizeM[ind,],col=ind,lwd=2,lty=ind)
}
legend("topright", rownames(Average_Latent_SizeM),lty=1:length(thenames), col = 1:length(thenames),cex=0.55,ncol=2)




meanAccAvgSize <-  colMeans(Average_Latent_SizeM[fastRows,])
meanAccAvgSize <-  rbind(meanAccAvgSize,colMeans(Average_Latent_SizeM[pearsonRows,]))
meanAccAvgSize <-  rbind(meanAccAvgSize,colMeans(Average_Latent_SizeM[spearmanRows,]))
meanAccAvgSize <-  rbind(meanAccAvgSize,colMeans(Average_Latent_SizeM[!T_BlindRows,]))
meanAccAvgSize <-  rbind(meanAccAvgSize,colMeans(Average_Latent_SizeM[T_BlindRows,]))
meanAccAvgSize <-  rbind(meanAccAvgSize,colMeans(Average_Latent_SizeM[corRankRows,]))
meanAccAvgSize <-  rbind(meanAccAvgSize,colMeans(Average_Latent_SizeM[maxCorRankRows,]))
meanAccAvgSize <-  rbind(meanAccAvgSize,colMeans(Average_Latent_SizeM[RLMfitMethod,]))
meanAccAvgSize <-  rbind(meanAccAvgSize,colMeans(Average_Latent_SizeM[!RLMfitMethod,]))

miny =1
maxy = 5*max(meanAccAvgSize)

plot(thr,meanAccAvgSize[1,],ylim=c(miny,maxy),
     main="Mean Size of Average-Latent-Variable",
     xlab="Correlation-Matrix's Maximum Goal",
     ylab="Size",
     type="l",
     col=1,
     lwd=2,
     log = "y")
for (ind in 2:nrow(meanAccAvgSize))
{
  lines(thr,meanAccAvgSize[ind,],col=ind,lwd=2,lty=ind)
}
legend("topright", legnames,lty=1:length(legnames), col = 1:length(legnames),cex=0.75)




miny = min(La_SignificantM)
maxy = 20*max(La_SignificantM)

plot(thr,La_SignificantM[1,],ylim=c(miny,maxy),
     main="Number of Discovered Features",
     xlab="Correlation-Matrix's Maximum Goal",
     ylab="Number of Features",
     type="l",
     col=1,
     lwd=2,
     log = "y")

for (ind in 2:nrow(La_SignificantM))
{
  lines(thr,La_SignificantM[ind,],col=ind,lwd=2,lty=ind)
}
legend("topleft", rownames(La_SignificantM),lty=1:length(thenames), col = 1:length(thenames),cex=0.55,ncol=2)




meanDiscovered <-  colMeans(La_SignificantM[fastRows,])
meanDiscovered <-  rbind(meanDiscovered,colMeans(La_SignificantM[pearsonRows,]))
meanDiscovered <-  rbind(meanDiscovered,colMeans(La_SignificantM[spearmanRows,]))
meanDiscovered <-  rbind(meanDiscovered,colMeans(La_SignificantM[!T_BlindRows,]))
meanDiscovered <-  rbind(meanDiscovered,colMeans(La_SignificantM[T_BlindRows,]))
meanDiscovered <-  rbind(meanDiscovered,colMeans(La_SignificantM[corRankRows,]))
meanDiscovered <-  rbind(meanDiscovered,colMeans(La_SignificantM[maxCorRankRows,]))
meanDiscovered <-  rbind(meanDiscovered,colMeans(La_SignificantM[RLMfitMethod,]))
meanDiscovered <-  rbind(meanDiscovered,colMeans(La_SignificantM[!RLMfitMethod,]))

miny = min(meanDiscovered)
maxy = max(meanDiscovered) + 10

plot(thr,meanDiscovered[1,],ylim=c(miny,maxy),
     main="Average Number of Discovered Features",
     xlab="Correlation-Matrix's Maximum Goal",
     ylab="Number of Features",
     type="l",
     col=1,
     lwd=2,
     log = "y")
for (ind in 2:nrow(meanDiscovered))
{
  lines(thr,meanDiscovered[ind,],col=ind,lwd=2,lty=ind)
}
legend("bottomright", legnames,lty=1:length(legnames), col = 1:length(legnames),cex=0.75)



SigDeM[is.na(SigDeM)] <- 0
miny = 1
maxy = 20*max(SigDeM)

plot(thr,SigDeM[1,],ylim=c(miny,maxy),
     main="Number of Significant Latent Variables",
     xlab="Correlation-Matrix's Maximum Goal",
     ylab="How Many",
     type="l",
     col=1,
     lwd=2,
     log = "y")

for (ind in 2:nrow(SigDeM))
{
  lines(thr,SigDeM[ind,],col=ind,lwd=2,lty=ind)
}
legend("topright", rownames(SigDeM),lty=1:length(thenames), col = 1:length(thenames),cex=0.55,ncol=2)



SigLatent <-  colMeans(SigDeM[fastRows,])
SigLatent <-  rbind(SigLatent,colMeans(SigDeM[pearsonRows,]))
SigLatent <-  rbind(SigLatent,colMeans(SigDeM[spearmanRows,]))
SigLatent <-  rbind(SigLatent,colMeans(SigDeM[!T_BlindRows,]))
SigLatent <-  rbind(SigLatent,colMeans(SigDeM[T_BlindRows,]))
SigLatent <-  rbind(SigLatent,colMeans(SigDeM[corRankRows,]))
SigLatent <-  rbind(SigLatent,colMeans(SigDeM[maxCorRankRows,]))
SigLatent <-  rbind(SigLatent,colMeans(SigDeM[RLMfitMethod,]))
SigLatent <-  rbind(SigLatent,colMeans(SigDeM[!RLMfitMethod,]))

miny = 1
maxy = max(SigLatent) + 10


plot(thr,SigLatent[1,],ylim=c(miny,maxy),
     main="Average # of Significant Latent Variables",
     xlab="Correlation-Matrix's Maximum Goal",
     ylab="How Many",
     type="l",
     col=1,
     lwd=2,
     log = "y")
for (ind in 2:nrow(SigLatent))
{
  lines(thr,SigLatent[ind,],col=ind,lwd=2,lty=ind)
}
legend("bottomright", legnames,lty=1:length(legnames), col = 1:length(legnames),cex=0.75)




sparcityM[is.na(sparcityM)] <- 0
miny = min(sparcityM)
maxy = max(sparcityM) + 0.75 

plot(thr,sparcityM[1,],ylim=c(miny,maxy),
     main="Matrix Sparcity",
     xlab="Correlation-Matrix's Maximum Goal",
     ylab="Sparcity",
     type="l",
     col=1,
     lwd=2,
     log = "y")

for (ind in 2:nrow(sparcityM))
{
  lines(thr,sparcityM[ind,],col=ind,lwd=2,lty=ind)
}
legend("topright", rownames(sparcityM),lty=1:length(thenames), col = 1:length(thenames),cex=0.55,ncol=2)




meanSparcity <-  colMeans(sparcityM[fastRows,])
meanSparcity <-  rbind(meanSparcity,colMeans(sparcityM[pearsonRows,]))
meanSparcity <-  rbind(meanSparcity,colMeans(sparcityM[spearmanRows,]))
meanSparcity <-  rbind(meanSparcity,colMeans(sparcityM[!T_BlindRows,]))
meanSparcity <-  rbind(meanSparcity,colMeans(sparcityM[T_BlindRows,]))
meanSparcity <-  rbind(meanSparcity,colMeans(sparcityM[corRankRows,]))
meanSparcity <-  rbind(meanSparcity,colMeans(sparcityM[maxCorRankRows,]))
meanSparcity <-  rbind(meanSparcity,colMeans(sparcityM[RLMfitMethod,]))
meanSparcity <-  rbind(meanSparcity,colMeans(sparcityM[!RLMfitMethod,]))

miny = min(meanSparcity)
maxy = max(meanSparcity)+0.25


plot(thr,meanSparcity[1,],ylim=c(miny,maxy),
     main="Mean Matrix Sparcity",
     xlab="Correlation-Matrix's Maximum Goal",
     ylab="Sparcity",
     type="l",
     col=1,
     lwd=2,
     log = "y")
for (ind in 2:nrow(meanSparcity))
{
  lines(thr,meanSparcity[ind,],col=ind,lwd=2,lty=ind)
}
legend("topright", legnames,lty=1:length(legnames), col = 1:length(legnames),cex=0.75)

1.2.2 Saving All


save.image("~/GitHub/LatentBiomarkers/RMD/DARWIN_ALL_Options.RData")